Publication number | US8126834 B2 |
Publication type | Grant |
Application number | US 12/427,037 |
Publication date | Feb 28, 2012 |
Filing date | Apr 21, 2009 |
Priority date | Apr 21, 2009 |
Fee status | Paid |
Also published as | US20100268678 |
Publication number | 12427037, 427037, US 8126834 B2, US 8126834B2, US-B2-8126834, US8126834 B2, US8126834B2 |
Inventors | Gao Chen, Claire M. Bagley |
Original Assignee | Oracle International Corporation |
Export Citation | BiBTeX, EndNote, RefMan |
Patent Citations (11), Non-Patent Citations (14), Classifications (5), Legal Events (3) | |
External Links: USPTO, USPTO Assignment, Espacenet | |
One embodiment is directed generally to a computer system, and in particular to a constraint based computer system that solves dynamic constraint satisfaction problems.
Many of the tasks that are addressed by decision-making systems and artificial intelligence systems can be represented as constraint satisfaction problems (“CSP”s). In this representation, the task is specified in terms of a set of variables, each of which can assume values in a given domain, and a set of constraints that the variables must simultaneously satisfy. The set of variables, domains and constraints is referred to as a CSP. Each constraint may be expressed as a relation, defined over some subset of the variables, denoting valid combinations of their values. A solution to a CSP is an assignment of a value to all the variables from their respective domains that satisfies all of the constraints.
A constraint based system includes a constraint solver that attempts to find one or more solutions to a given CSP, or prove that no solution exists. Constraint based systems are used for many artificial intelligence related applications and a variety of other applications, including: (1) Product configurators; (2) Robotic control; (3) Temporal reasoning; (4) Natural language processing; (5) Spatial reasoning; (6) Test-case generation for software and hardware systems; (7) Machine vision; (8) Medical diagnosis; (9) Resource allocation; and (10) Frequency allocation.
The network of constraints in a CSP can be viewed as a graph, having a node for each variable and an “arc” for each constraint. The members of each arc are the variables that appear in the constraint to which the arc corresponds. An arc is said to be consistent if for any variable of the arc, and any value in the domain of the variable, there is a valid assignment of values to the other variables on the arc that satisfies the constraint represented by the arc.
Classes of problems exist which are comprised of very large sets of variables that may only be conditionally related or required for a solution. One example of such problems is the configuration of large component-based systems. For example, selecting a type of hard disk controller for a computer configuration is not needed if a hard disk has not been chosen as a form of storage. If instead flash memory is chosen, a different set of variables and constraints would be required to be solved. Known CSP solvers do not allow the representation of conditional structure or reasoning over an inclusion of a variable in a solution. Techniques have been developed to allow such large problems to be represented as a set of smaller sub-problems, conditionally related through composition or association. A “dynamic constraint satisfaction problem” is one in which these sub-problems of variables and constraints can be incrementally added as required, either explicitly or as a result of inference from the propagation of constraints.
One known approach to minimize large CSP problems is referred to as “Conditional CSP”, and includes the notion of a variable being active or inactive, as well as constraints to activate a variable. In this approach, a variable is only assigned a value in the final solution if it is active. Conditional CSP is limited in that it does not provide any significant space savings in large problems, nor does it allow for segmentation of related variables into sub-problems. Another known approach is referred to as “Generative CSP” and extends Conditional CSP by introducing the concept of components, which are groups of related variables, and component type, which is the further extension and specialization of these components. However, similar to Conditional CSP, Generative CSP is still implemented in terms of activity state and does not provide real space savings.
One embodiment is a dynamic constraint solver system for solving a constraint satisfaction problem model that includes a plurality of ports. The system defines a hierarchical union that includes all problems in a lower port that is in a problem under another port in the model. The system generates a constraint that computes a cardinality of the hierarchical union and determines an included set and an excluded set for the hierarchical union. The system then propagates the included set and excluded set to participating ports of the hierarchical union.
One embodiment is a dynamic constraint satisfaction problem solver that implements a hierarchical union operator and corresponding constraint for a constraint satisfaction problem. The hierarchical union determines the union of all instances in a port within all instances of another port.
Computer readable media may be any available media that can be accessed by processor 22 and includes both volatile and nonvolatile media, removable and non-removable media, and communication media. Communication media may include computer readable instructions, data structures, program modules or other data in a modulated data signal such as a carrier wave or other transport mechanism and includes any information delivery media.
Processor 22 is further coupled via bus 12 to a display 24, such as a Liquid Crystal Display (“LCD”), for displaying information to a user. A keyboard 26 and a cursor control device 28, such as a computer mouse, is further coupled to bus 12 to enable a user to interface with system 10.
In one embodiment, memory 14 stores software modules that provide functionality when executed by processor 22. The modules include an operating system 15 that provides operating system functionality for system 10. The modules further include a dynamic constraint solver module 16 that performs dynamic constraint solving for models using hierarchy union operators as disclosed in more detail below. System 10 can be part of a larger system that includes a constraint solver, such as a product configurator or artificial intelligence system. Therefore, system 10 will typically include one or more additional functional modules 18 to include the additional functionality.
Each problem is formed of zero or more non-structural variables 206. Examples of non-structural variables 206 includes Boolean variables, integers, floating point variables, etc. Each problem 204 may also include zero or more structural variables or “ports” 208. A port is a container for problems and connects sub-problems to the problem or to another sub-problem or acts as an extension point from one problem to another problem. Each port 208 can be connected to zero or more sub-problems 204. A port may be defined by two items: (a) the definition of the problem to be connected to the port; and (b) a numeric domain representing how many instances of the problem is required or allowed in the port (referred to as the port's “cardinality”).
For example, a problem definition for problem A may be as shown in Example 1 below (the bracketed information indicates the domain for the problem/port):
ProblemA | ||
|_Port to ProblemB [0..5] | ||
|_Resource [1..10] | ||
As shown in the definition, Problem A includes a port to Problem B. According to that port, Problem A may include zero to five Problem Bs. Each Problem B is defined with an integer amount of Resource [1 . . . 10] that it can provide to a resource sum, which is the sum of all resources of Problem Bs that are connected to the port to Problem B. A resource sum constraint is further disclosed in pending U.S. patent application Ser. No. 12/362,209, filed on Jan. 29, 2009, and herein incorporated by reference. The cardinality domain for the port to Problem B is [0 . . . 5].
In one embodiment, a “hierarchical union” operator represents all problems in a lower port that is in a problem under another port. Example 2 below illustrates a model that a hierarchical union can be applied to:
ProblemA | ||
|_Port1 (to ProblemB) [0..4] | ||
|_Port2 (to ProblemC) [0..5] | ||
|_Resource [1..10] | ||
A hierarchical union operator “Port1.HierarchicalUnion(Port2)” represents the collection of all Problem Cs under Problem Bs that are under Problem A.
In one embodiment, the hierarchical union is used to reason over variables that are aggregated over a nested hierarchy of potential dynamic instances. For example, using Example 2, the hierarchical union can be used to aggregate the resource variable over all Problem Cs under Problem A through Problem B using: Port1.HierarchicalUnion(Port2).sum(Resource). This hierarchical union provides all Problem Cs under Problem A through Problem B so that a resource sum constraint can collect all resources over these Problem Cs.
A port variable represents a collection of problems. Its domain composes the port's cardinality, a collection of problems that are already included in the port (referred to as its “included set”), a collection of problems that are already excluded from the port (referred to as its “excluded set”), as well as candidate problems that are neither excluded nor included in the set at that moment.
As an example, consider the following model (Example 3):
ProblemA | ||
|_Port1 (to ProblemB) [0..5] | ||
In one embodiment, a port is “closed” when it cannot take any new candidates, or its candidates are restricted to a known set of instances. This can happen when a port is fully bound, or when it is a subset of another port that is closed. When a port is bound, the number of problems the port can have in the final solution is already determined as the bound cardinality, and all these problems have been determined since they are already in the port. The port has no room for any new problems and hence is closed.
If a port A is a subset of a port B which is closed, port B cannot have any new candidates. It can be implied that port A also cannot have any new candidates, and therefore port A is also closed. For example, consider a port “P1” defined with a cardinality [2 . . . 2] and a port “P2” defined with cardinality [0 . . . 4]. P1 and P2 are the same type of ports, meaning they can contain instances of problems of the same type definition. Consider also a constraint that states that P1 is a subset of P2. Assume P2's cardinality is bound to the value 3 and contains 3 sub-problems: P2#1, P2#2 and P2#3. P2 is closed since it is bound, and P1 it also closed since it is a subset of a closed port, but P1 is not bound. P1's cardinality is bound to 2, but there are 3 candidates for 2 “spots” in this port (e.g., any combination of 2 problems from the set {P2#1, P2#2, P2#3}). Although it cannot be decided for now which two will be finally selected for the solution, P1 is closed since it cannot take any new candidates.
In one embodiment, the hierarchical union represents all problems in a lower port that is in a problem under another port, and therefore it is a collection of problems itself. Consequently, a hierarchical union can also be considered a port having its own cardinality, included set and excluded set.
As an example of a hierarchical union as a port, consider the model in Example 4 below:
ComponentA | ||
|_Port1 (to ComponentB) [0..4] | ||
|_Port2 (to ComponentC) [0..5] | ||
For the hierarchical union constraint of “Port1.HierarchicalUnion(Port2)”, another constraint can be generated internally to compute the cardinality of the hierarchical union: “HierarchicalUnion.Cardinality=Port1.sum(Port2.Cardinality)”. The right hand side of this constraint is a resource sum constraint that sums up the cardinality of all Port 2 on Problem Bs in Port 1.
Referring again to the instantiation of
For simplicity purposes, the above example illustrates a case where the cardinality of each port is bound to a value. In contrast, in general, embodiments compute the domain of the cardinality which includes a lower bound and an upper bound. This internal constraint can reduce information in all directions. For example, it can reduce the domain of the hierarchical union's cardinality. It can reduce the domain of Port 1's cardinality. It can also reduce the domain of Port 2's cardinality on a Problem B in Port 1.
In one embodiment, solver 16 computes the included set of the hierarchical union. As illustrated in
In one embodiment, solver 16 propagates the excluded set to participating Port 2s. If, for some reason, a Problem C is excluded from the hierarchical union, then solver 16 knows that this Problem C must also be excluded from any Port 2 on a Problem B in Port 1. Therefore, solver 16 keeps track of all Port 2s on the problems in Port 1. It iterates through all Port 2s, and excludes all problems in the hierarchical union's excluded set from each Port 2 by adding them to the Port 2's excluded set.
In one embodiment, solver 16 then computes the excluded set. The excluded set for the hierarchical union can be computed when Port 1 is closed such that Port 1 can have no new Problem Bs in it. Otherwise, if Port 1 can still have a new Problem B later, nothing can be excluded from the hierarchical union since the Port 2 under the possible new Problem B may include anything. Solver 16 keeps track of all Port 2s on the problems in Port 1. If Port 1 is closed, solver 16 iterates through all Port 2s, looking for problems that are excluded from all Port 2s. It then excludes these problems from the hierarchical union by adding them to the hierarchical union's excluded set.
In one embodiment, solver 16 propagates the included set to a participating Port 2. When a Problem C is included in the hierarchical union, it must be included in at least one Port 2 on a Problem B in Port 1. If Port 1 is closed, it should have all its Problem Bs already as in the final solution. If the Problem C is excluded from every Port 2 on a Problem B except for one, then solver 16 includes the Problem C in the only Port 2 that has not excluded it yet.
At 1002, a hierarchical union is defined for a dynamic CSP that includes problems and sub-problems connected by ports. The hierarchical union represents all problems in a lower port that is in a problem under another port.
At 1004, a constraint is generated that computes the cardinality of the hierarchical union.
At 1006, the included set of the hierarchical union is computed.
At 1008, the excluded set of the hierarchical union is computed.
At 1010, the excluded set and the included set of the hierarchical union is propagated to participating ports.
As disclosed, an embodiment is a solver for a dynamic CSP that defines a hierarchical union and allows a constraint to be defined over the hierarchical union. The hierarchical union includes a cardinality, an included and an excluded set that is propagated to affected ports.
Several embodiments are specifically illustrated and/or described herein. However, it will be appreciated that modifications and variations of the disclosed embodiments are covered by the above teachings and within the purview of the appended claims without departing from the spirit and intended scope of the invention.
Cited Patent | Filing date | Publication date | Applicant | Title |
---|---|---|---|---|
US7069537 | Aug 19, 2003 | Jun 27, 2006 | Oracle International Corporation | Techniques for partial loading of a configuration associated with a configuration model |
US7333968 | Aug 17, 2005 | Feb 19, 2008 | International Business Machines Corporation | Conditional CSP solving using constraint propagation |
US7515156 * | Jan 8, 2003 | Apr 7, 2009 | Hrl Laboratories, Llc | Method and apparatus for parallel speculative rendering of synthetic images |
US8065255 * | Nov 13, 2008 | Nov 22, 2011 | Oracle International Corporation | Management of sub-problems in a dynamic constraint satisfaction problem solver |
US20020107749 | Feb 5, 2001 | Aug 8, 2002 | David Leslie | Networked based paralleling switchgear equipment configuration process |
US20020143653 | Dec 29, 2001 | Oct 3, 2002 | Dilena Ettore | Configuration system and methods |
US20020166089 | Mar 6, 2001 | Nov 7, 2002 | Amos Noy | System and method for test generation with dynamic constraints using static analysis |
US20040019852 | Jul 23, 2002 | Jan 29, 2004 | Xerox Corporation | System and method for constraint-based document generation |
US20050198621 | Mar 2, 2004 | Sep 8, 2005 | Microsoft Corporation | Efficient checking of state-dependent constraints |
US20050278271 | May 13, 2005 | Dec 15, 2005 | Anthony James T | System and method for determining a product configuration |
US20070094184 | Aug 17, 2005 | Apr 26, 2007 | International Business Machines Corporation | Solving constraint satisfaction problems with duplicated sub-problems |
Reference | ||
---|---|---|
1 | Benhamou, Frédérick et al., "Applying Interval Arithmetic to Real, Integer and Boolean Constraints", Journal of Logic Programming, 32(1), 1997. | |
2 | Fleischanderl, Gerhard et al., "Configuring Large Systems Using Generative Constraint Satisfaction", IEEE, pp. 59-68, No. 1094-7167/98, Jul./Aug. 2008, IEEE. | |
3 | Frühwirth, Thom et al., "Principles of Constraint Systems and Constraint Solvers", Archives of Control Sciences: Special Issue on Constraint Programming, 16(2) http://www.informatik.uni-ulm.de/pm/mitarbeiter/fruehwirth/Papers/acs-systems3.pdf, 2006. | |
4 | Gelle, Esther et al., "Solving Methods for Conditional Constraint Satisfaction", IJCAL, 2003. | |
5 | * | Grid workflow recovery as Dynamic constraint satisfaction problem, Dragiev, S.; Schneider, J.; Open Systems (ICOS), 2010 IEEE Conference on Digital Object Identifier: 10.1109/ICOS.2010.5720067 Publication Year: 2010 , pp. 74-79. |
6 | ILOG, "ILOG Configurator, Powering online product and service configuration applications", Product Datasheet, Mar. 2005, ILOG.S.A. | |
7 | Lhomme, Olivier, "Consistency Techniques for Numeric CSPs", Proceedings in IJCAI-93 pp. 232-238, 1993 Chambery, France. | |
8 | Mittal, Sanjay et al., "Dynamic Constraint Satisfaction Problems", Proceedings of the Eighth National Conference on Artificial Intelligence, 1996. | |
9 | Oracle, "Oracle Configurator", Oracle Data Sheet, 2008, Oracle. | |
10 | Sabin, Daniel et al., "Configuration as Composite Constraint Satisfaction", Proceedings of the (1st) Artificial Intelligence and Manufacturing Research Planning Workshop, 1996. | |
11 | Selectica, "Selectica Configuration", Solution Overview, 2005, Selectica, Inc., San Jose, CA. | |
12 | Stumptner, Markus et al., "Generative Constraint-Based Configuration of Large Technical Sytems", Artificial Intelligence for Engineering Design, analysis and Manufacturing, pp. 307-320, No. 12, 1998, Cambridge University Press, USA. | |
13 | Tacton, "Tacton's Sales Configurator-Bridging the gap between engineering and sales for selling customized products.", http://www.tacton.com/templates/page-68.aspx?epslanguage=EN, 2007, Tacton Systems AB, Sweden. | |
14 | Tacton, "Tacton's Sales Configurator—Bridging the gap between engineering and sales for selling customized products.", http://www.tacton.com/templates/page—68.aspx?epslanguage=EN, 2007, Tacton Systems AB, Sweden. |
U.S. Classification | 706/46 |
International Classification | G06N5/02, G06F17/00 |
Cooperative Classification | G06N5/04 |
European Classification | G06N5/04 |
Date | Code | Event | Description |
---|---|---|---|
Apr 21, 2009 | AS | Assignment | Owner name: ORACLE INTERNATIONAL CORPORATION, CALIFORNIA Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:CHEN, GAO;BAGLEY, CLAIRE M.;REEL/FRAME:022571/0167 Effective date: 20090416 |
May 29, 2012 | CC | Certificate of correction | |
Aug 12, 2015 | FPAY | Fee payment | Year of fee payment: 4 |